To Diagnose and Classify Plant Diseases, A Machine Learning System was Deployed
Naveenkumar C U M1, Vidhya S2
Dept of MCA, Bangalore Institute of Technology, Bangalore, India1
Dept of MCA, Asst Professor, Bangalore Institute of Technology, Bangalore, India2
ABSTRACT
Due to the country's burgeoning population and severe hunger, agriculture plays a crucial role in Indian society. Therefore, more harvest output is required. Germs, infections, and other organisms are a major cause of lower harvest yield. Researching and understanding plant diseases is a crucial part of farming. Plant diagnosis techniques are often used beforehand. It's challenging to keep track of plant illnesses, make observations, and apply treatments manually. To be able to save time and effort, image processing is utilized to distinguish between various plant diseases.Machine learning methods may be utilized to classify plant diseases; these methods include creating a dataset, importing images, doing pre-processing and segmentation, extracting features, training a classifier, and classifying the data.
The fundamental objective the goal of this study became to develop an algorithm that can recognise a healthy lifestyle unhealthy harvest leaves and predict plant diseases. Using a publicly available dataset consisting of 54,306 images of sick and healthy plant leaves obtained under controlled circumstances, the authors of such studies taught a computer to recognize some distinct harvests and 26 diseases. This article makes use of the ResNets algorithm. An example of an artificial neural network is the residual neural network (ResNet). To combat the vanishing/exploding gradient issue, the ResNet method provides a residual block.Residual Networks may also be constructed using the ResNetmethod.When it comes to labeling images, ResNets do very well. ResNets methods made use of a number of settings, including learning rate scheduling, gradient clipping, and weight decay.
Cyberbullying, Machine learning, Natural language processing, Social media.
Machine learning algorithm, image recognition, predictive models, predictive algorithm, classification algorithm, task analysis, Diseases